Vidchecker products are designed for testing media files, i.e. video and/or audio files, to a specific set of technical parameters, as defined in a test template.
Examples are checking the resolution and frame rate of the video, the video levels, the audio levels and loudness.
Uniquely, Vidchecker products can also automatically correct common errors in video and audio, saving the time-consuming cycle of returning a file for manual correction then doing the QC again (perhaps a few times over, as manual correction can be error-prone)
Vidchecker & Vidchecker-post are Windows software applications to provide automatic quality control and optional correction of file-based media.
Vidchecker and Vidchecker-post have all the same test and correction features EXCEPT that:
As the name suggests, Vidchecker-post is designed for post-production companies or those with a lower volume of content, where checking 1 file at a time is sufficient.
Vidchecker does a comprehensive check of all aspects of media files, covering:
Some examples with fixing option:
If you are unsure and wish to save money, you can start with Vidchecker-post and upgrade to Vidchecker, or start with Vidchecker and upgrade to the additional correction option.
Vidchecker always leaves the original input file as it is, with no changes. If a corrected file is produced by Vidchecker this is a new file, with a different filename (typically the original filename plus ‘corr’ with the date and time, although other suffixes can be added).
Actually, the Vidchecker team designed and produced the first product in this area (now sold by Tektronix). We have been working in this area since 2003 and are probably the most experienced team writing this type of software.
Consequently, Vidchecker products have much newer software technology – as one example, automatically using all the processor cores for faster processing – but at the same time based upon many years of experience allowing us to provide a product much more suited to user needs – such as making it the easiest product to use – and to implement features quickly (as we have done most of them before).
Many existing users of other products are now converting to use Vidchecker products.
Vidchecker is the 2nd generation of auto QC software, based upon our years of experience and latest software technology.
Vidchecker does thorough tests of the syntax, file, video and audio parameters (actually, doing similar or superior tests to competing products).
Vidchecker automatically using all processor cores (up to 5x faster than competing products), without using GPUs.
Also, unlike competing products Vidchecker does not require files to be copied locally.
Vidchecker can optionally correct the common video and audio problems (incl. loudness) and write out a full quality, corrected file (re-encoded if required).
Vidchecker has been designed with our years of experience to be easier to use by operators and others who are new to file-based QC – which is most people.
As just one example: some competing products show all instances of an error – which is fine unless the same error occurs in most frames of a movie with 100,000 frames, which then generates a report 100 pages long, which no-one reads – but buried in the middle is a critical error. You can see all the errors with Vidchecker if you wish, but typically are limited to show (for example) only the details of the first 10 instances of an error with a count of how many more errors of the same type there are. So in this example, Vidchecker gives a report just 1/2 page long – and the other critical error is easily seen.
A second example: Vidchecker has a syntax check turned on as standard – there is no need to switch it on explicitly (unlike competing products). This standard syntax check is intelligent in that it is a check of the sort ‘will this file likely play correctly in standard players, even if the syntax is not 100% correct’.
If you wish to do a detailed syntax check there is a single test to turn on ‘enhanced syntax checking’, which does a more strict check.
If competing products find a syntax error they may report something like ‘RBSP_trailing_bits missing’ [an actual example] – which of course is meaningless to an operator.
Whereas if Vidchecker find a syntax error then the report is ‘syntax error’ and the timecode is given – which we think is much more useful.
There are many such examples with how Vidchecker is designed, and many customers buy our products because of the ease of use.
Vidchecker offers an easy click-through Windows installation.
Both products have a comprehensive, well documented API. Customers who have integrated Vidchecker have told us that the integration time is typically 3x faster with our products than competing products.
Vidchecker uses standard hardware and does not need GPUs to achieve optimal processing speed. This means that Vidchecker can easily be moved to different servers, and can by run in virtualised environments such as Amazon Web Services.
The range of products include:
You can easily upgrade from Vidchecker-post all the way to a Grid, and for some customers who are new to file-based QC – even those with a greater file throughput – Vidchecker-post is a good low-cost way to get an excellent understanding of file-based QC with a full-featured product.
Many Vidchecker features are added at a customers request – we genuinely wish to receive requests from customers and we are often able to incorporate the requested features quickly.
Just fill in the form and we will send you the download details.
We recommend that prior to using the download that you have an online demo, as our experience is that this is the fastest way to get going with the software – see below.
The trial version will auto-license for 15-days.
Vidchecker has all the codecs enabled for decode and re-encode, although re-encoded video has a watermark added and audio is time-restricted.
The trial can easily be extended if needed – just ask us.
This typically takes 45 minutes for a good overview, plus some time for questions and looking at specific tests that might be of interest to you. Our experience is that this is the fastest way to get going with the software. We can of course provide a demo outside of standard business hours by arrangement (contact sales for information).
There are several videos of our products, and others available on YouTube. Some of these videos have been done by ourselves, some by resellers or partners, and are being added to over time.
Please go to Youtube.com and search for ‘Vidchecker’.
Vidchecker is designed to remove the labor-intensive tasks of checking conformance by hand and checking video/audio compliance with waveform monitors and audio loudness meters, and manually correcting levels with video processing amplifiers, and other tools.
There are many file errors which are not actually visible e.g. GOP structure, VBR/CBR, video profile/level, bit-rate. which may be crucial to a file playing out correctly (or not).
A ‘not uncommon’ workflow using software that does auto QC only can be as follows:
.. and so back to step 1; steps 1 and 2 are repeated
And now the file passes the video levels test.
BUT fails the audio loudness test!
So now steps 3 and 4 are repeated.
And while there the editor notices that a transition between two segments is not so good, so moves the edit point by a few frames .. not noticing that the edit points don’t exactly line up and accidentally there is a single black frame introduced into the final render.
Steps 1 and 2 are then repeated – again – and yet again the file has to be sent back for edit.
This third time around the loop of ‘QC then manual correction’ the file finally passes QC – after much manual time spent, doing the QC 3 times, etc.
So after all this time:
By far the most common errors are audio levels & loudness, and video levels/gamut.
Vidchecker corrects these automatically, allowing the above shorter, faster workflow to occur.
For details of the video & audio level corrections done by Vidchecker see below.
And when Vidchecker does the correction, a new file is produced – the original input file is always left unaltered.
For the chroma/luminance tests the default values can be used (typically the recommended EBU limits) or custom values easily set. Options can be selected such as testing to EBU R.103.
Vidchecker does NOT just do legalisation – for details of the video correction and the affect on video quality see below.
Vidchecker does NOT do legalisation – doing legalisation can be visible, particularly on bright areas which are then ‘crushed’ to within the legal range.
Vidchecker uses unique patented algorithms for video correction that we call ‘intelligent correction’, designed to correct video levels in such a way that the changes are not visible to the human eye.
At Telestream we believe that automated video correction is vitally important to speed up workflows as errors in video levels are some of the most common errors that occur – and correcting these by hand in an NLE is time-consuming, expensive, and can be an error-prone.
So in 2009 Vidchecker engineers spent several months researching how the human eye/brain perceive static and moving images and how we could do corrections without impairing the video quality.
The corrections that Vidchecker does vary over time, based upon the video content itself. The software intelligently checks the video changes spatially and temporally (hence the name ‘intelligent correction’), in both luminance and chrominance and modifies areas of the video – both those outside the correct tolerances and within the legal ranges – in such a way so that the changes are not perceivable to humans. (We have done ‘double-blind’ tests, where we have asked experienced video QC personnel to compare two videos side-by-side on the same screen, to determine which of the two videos is the original and which is the corrected – and we have never found anyone who can discern a difference.)
Telestream has patented these algorithms, and no other company is authorised to use these.
The result is that unlike any other product, using Vidcheck’s correction algorithms maintains the video quality and makes auto correction of video feasible.
Vidchecker decodes the video for correction and then re-encodes it either using it’s built-in software encoders, or by passing it via a watch folder to existing encoders in your workflow. Pricing is available with and without built-in encoders to re-encode the file after correction. Please note that there are some limitations in the formats available from the built-in re-encoding.
The audio correction done is scaling (up or down, as required), after measurement in accordance with the appropriate standard (e.g. BS-1770-3).
Custom levels can be set on all these tests.
This is fine for AC-3 / E-AC3 which allows accompanying metadata to be broadcast and control the consumer’s decoder, but the majority of broadcast audio (MPEG1/2, PCM, AAC etc) does not have this facility. The loudness has to be checked and controlled before it is broadcast.
For the full current list please see the relevant product pages, but a summary is:
Corrected files will be output In the same format as the Input file. If format conversion is required as part of a workflow, Vidchecker integrates seamlessly with Telestream’s Vantage product allowing users to transcode their corrected file to a diverse range of formats.
The audio-only file can be in any of the supported audio formats (either in a wrapper/container or as ‘essence’ only).
There is no limitation. An audio track can be mono, stereo, 5.1 or 7.1 and you can have as many audio tracks as you wish (for example, one Vidchecker user in London produces files with 24 stereo audio tracks, for different European languages).
Yes, easily using the ‘channel mapping’ function.
In this example, the channel mapping function allows you to define
The video essence file can be in any of the supported video formats.
Vidchecker has a very well-defined and well-documented API, and we know that all the functions work as our own GUI uses the API.
It is already directly integrated with most of the leading third party applications – for example
Telestream Vantage, Amberfin iCR, Digital Rapids Transcode Manager, Sony MediaBackbone, Root6 ContentAgent, R&S DVS, Envivio, Dalet, Suitcase TV and many other leading workflow and transcode systems.
Please see our integration page for the current list – but note that this list is rapidly growing so if the third party system is not listed it may be that it is already currently being worked on – please ask us.
There are two options:
Using watch folders is very quick and easy: your system moves a file to be tested into a folder, Vidchecker automatically detects the arrival of the file, tests it and then moves the file to pass / reject / quarantine folders and your system detects the move.
Using the Vidchecker SOAP API is straightforward as it is comprehensive and well documented (our own GUI uses the API, so we know it all works!) and many customers have done this.
The API controls include
The full API documentation is available from Telestream under NDA – please contact firstname.lastname@example.org to request this.
There can be several reasons:
For the first two reasons, on many occasions we can help via an online session, or you can send us sample files for us to configure a test Template where we send you the details of the set up required to find the fault.
Vidchecker is now very good at finding visual and audio artefacts (typically better than competing products) but there may be visual / audio faults that the software does not reliably find.
We are always eager to improve our product, so if you are able to provide test samples of video files then we really want to hear from you and we may be able to incorporate the test in a future version.
Humans are good at seeing what is a visual defect and what should be in a scene, with all our contextual knowledge and understanding particularly if single-stepping through the video frame-by-frame.
For example, in a scene with a man standing in front of a checkerboard patterned wall, wearing a highly patterned tie, perhaps a single macroblock is out of place or has moved, in a single frame (a real example!). Even to a person this defect is not obvious (and takes several seconds of careful scrutiny of each frame), so software will find it very hard or impossible to detect reliably.
Conversely, if it is that hard for people to see such a defect, how many ‘real’ viewers will see this defect?
And compare the cost of doing a very fast, reliable, consistent, much lower cost automated QC versus trying to do this ‘several seconds per frame expert human QC’.
Sometimes we have been asked for the software to find such ‘defects’ as:
These are actual requests, where we have been asked why the software does not report these ‘defects’.
Regarding the skin tone, the software has no way of knowing if it is a sunset scene, or the person is meant to be an alien with red skin, or there is a red neon light nearby, or..
For the scenes with black areas, likewise
And for the audio, the software does not know if the talking is meant to be whispering, or meant to be out of earshot of other people in the scene, or the person is losing their voice, or..
We consider that all these types of issues are really editorial or artistic decisions – not technical quality checks.
Our software is great for detecting technical issues, but cannot check the editorial or artistic choices that have been made.
A single PC server license of Vidchecker enables up to 4 files to be processed at the same time (one file at a time using Vidchecker-post). The software is optimized to make use of recent multi-core and muti-thread CPUs to maximise processing speed for larger video files. Additional licenses can be added to the same PC server to test 8, 12, 16 or more files at the same time or installed on other servers. It can also be scaled up to use multiple servers as a Vidchecker Grid (see below).
So there are many variables!
But, to give an approximate guide:
Then 4 hours of HD video will be processed in 4 hours.
Please bear in mind that this could run 3 or more times slower if all the tests are turned on, or could run 5 times faster if really fast 24-core processors are used.
We always recommend that when you have a trial of Vidchecker that you check the processing, using a representative set of tests turned on, running on representative hardware and checking for any bottlenecks that are slowing the processing.
This is a deliberate choice, for two reasons:
Multiple networked PC servers with Vidchecker licenses can work together as a Vidchecker Grid for enterprise installations requiring very large file throughputs. The ‘primary node’ server handles all the control and interface, manages the workload and hands out processing tasks to the ‘secondary node’ servers.
There are 3 selectable levels of checking. Checks in the test Templates can be set so that if the check does not pass, Vidchecker will:
Vidchecker is supplied with standard default test Templates that are good to start with. For each test, the default values provided give a good basis to start with and then you can customize them if needed.
For example, there are Templates for
In addition, you can use the ‘Auto Template’ feature. This uses a file that you supply, that you know is ‘good’ (i.e. it passes your current QC tests). This known good file is examined by Vidchecker which then automatically sets up and configures a test Template to match, switching on commonly selected tests (e.g. for video levels). This does not do everything, but is a good starting point.
Single or multiple files can be
The user switches on the tests in the test Templates. These include:
Although there are some common standards, e.g. for loudness measurement, or common values for video resolution, there are many parts of the configuration of a media file which vary from one broadcaster / content distributor / post house to another, such as number of audio tracks used, whether color bars or black sections are included, or captions included, GOP structures, bit rates, acceptable visual quality levels.
Even a single broadcaster will have many different Templates e.g.
Commonly, a Vidchecker user may have from a few to a few hundred Templates (in the latter case, normally one Template for each content provider).
There is no limit (other than disk space: a single test Template requires less than 20KB in the Vidchecker database).
Once the checking is finished, the user is presented with a screen summarizing the results of the checking (the Results screen). An integrated media player enables you to click and play the video at the alert points
Reports can also be sent to locations defined in the test Templates:
All results are stored in the results database so that they can be searched and retrieved at a later date, and you have a complete audit trail of all your QC.
There are a number of reasons for this:
Vidchecker has two parts to the software:
It is the server application which is licensed, and controlled via a software-based license (see below).
The web-based GUI can run on the same PC as the Vidchecker server application is installed, and/or as many other PCs or Macs, Android, Linux devices.. as desired.
So for example, with a single Vidchecker license running on a PC in a server room testing and correcting files, you could have 10 or 20 different people on your network using the web-based GUI on their PCs / Macs to set up processing Tasks, view processing results, etc.
No dongle is required. Vidchecker uses a software-based licensing system. In order to Install Vidchecker you will need to obtain a license key from Telestream, the key comes in the form of a serial number that is sent to you via e-mail. Telestream can provide a time limited trial license key to allow you to evaluate the software. If you choose to buy the software then you will be issued a permanent license key.
The license can also be easily moved by the user to a new PC from an old PC without involving Telestream – see below.
When you install Vidchecker, the Vidchecker License Tool Is also installed, It can be found In the Vidchecker folder of the Windows start menu. To move your license you simply use the tool to deactivate the license on the licensed machine, and install Vidchecker on the new machine with the same license key.
Vidchecker works with Microsoft Windows operating systems:
Either server OS is fine; for desktop installations Windows 7 (64-bit) is recommended as having the best performance.
Yes. Vidchecker does not use any ‘special’ OS calls and runs well under VMWare or virtual systems such as Amazon Web Services (we test large Vidchecker Grid installations on Amazon Web Services).
Not natively. However, Vidchecker will run in a Windows environment under VMware / Parallels / Mac Bootcamp etc.
Also bear in mind that often the QC that is done by Vidchecker is required to run as fast as possible, so normally it is better to dedicate a PC / server to this task. Therefore even in a facility where operators work withMac or Linux, it's common to use a PC running Windows dedicated to Vidchecker QC. Since the PC stands alone and uses a web interface, it does not matter what OS is running inside it.
Normally customers have a good local supplier of PCs / servers, often with preferential pricing, installation and support and it is not cost-effective for Telestream to supply the hardware.
Also a particular customer may use specific I/O boards e.g. for fibre or 10Gbit networking.
And in any event, no special hardware is required – Vidchecker will even run on a laptop – so there is no benefit in Telestream supplying the hardware.
Vidchecker will automatically spread the processing across all cores / threads so the more processing power the faster the QC and correction will be done (subject to other bottlenecks in the system). And processing video files is highly compute-intensive, as there is a lot of data to process.
Therefore a PC / server with one or more recent quad core i7 or hex core Xeon processors is recommended, with at least 16GB RAM for a Vidchecker-post or 32GB RAM for each Vidchecker licence.
Vidchecker does not require the media files to be copied locally before processing (unlike some other competing products), but it is vital that the PC / server has fast access to the media files – there is no point having a very fast PC if the PC is waiting to access the media.
Therefore fibre channel or 10GBit network access to the media is recommended
A fast local disk (RAID0) or SSD is recommended, particularly if correcting media files. The local disk would need to be of sufficient capacity to store temporary copies of uncompressed media, as Vidchecker does all corrections on uncompressed video / audio.
We recommend 2TB of fast RAID0 storage as a minimum.